中文摘要：

近年来,深度学习中的卷积神经网络已经广泛运用于图像识别领域,它不仅显著提升了识别准确率,同时在特征提取速度方面也优于许多传统方法。针对高速公路环境下的车型识别问题,引入卷积神经网络（CNNs）理论,设计相应特征提取算法,并结合SVM分类器构建识别系统。通过对高速公路上主要三种车型（小车、客车、货车）的分类实验显示,该方法在识别精度及速度上均取得了较显著的提高。

英文摘要：

In recent years,the deep convolution neural network（ CNN）,a state-of-the-art deep learning method,has been widely used in the field of image recognition. It can not only significantly improve the recognition accuracy,but also superior to many traditional algorithms in terms of feature extraction speed. This paper firstly introduced the CNN for the highway vehicle recognition. It constructed a vehicle recognition system by using a proposed deep CNN based feature extraction method and the SVM classifier. The classification results of three major types of vehicles（ cars,buses,trucks） on the highway show that significant improvements are achieved in both classification accuracy and speed.